WO1996019939A1

WO1996019939A1 - A device and a method for recording and monitoring cardiac activity signals

Info

Publication number: WO1996019939A1
Application number: PCT/BE1995/000123
Authority: WO
Inventors: Luc Hondeghem; Roland Stroobandt
Original assignee: Luc Hondeghem; Roland Stroobandt
Priority date: 1994-12-28
Filing date: 1995-12-28
Publication date: 1996-07-04
Also published as: EP0801539A1

Abstract

The electrical signal generated by a heart during its activity is detected and signal processing means are provided for extracting and processing from said electrical signal those frequency components which are above 500 Hz without removing them.

Description

"A device and a method for recording and monitoring cardiac activity signals"

The invention relates to a device for recording and monitoring cardiac activity signals, said device comprises detecting means provided for detecting an electrical cardiac activity signal generated by a heart during its activity.

Such a device is known from the US-A-4.947.857. The known device is used for detecting and analyzing electrocardiogram signal data for diagnostic purposes. The device comprises a set of electrodes which are applied to the body of a patient. The electrodes capture the electrical cardiac activity signal generated by the heart during its activity. The electrical signal is sampled and averaged and Fourier transformed to the frequency domain. The Fourier transformation and averaging characteristically attenuates all signals above 500 Hz, since they are not considered to comprise relevant information. The averaging of the sampled signal is indicative that weaker, high frequency signals are not considered. In monitoring cardiac activity signals the P, the QRS and the T wave are considered as comprising relevant information about the cardiac activity. Those waves generally are present in a frequency range of 0-40 Hz. The His-bundle, which is situated in the 30-250 Hz also comprises relevant information. The article "Standards for analysis of ventricular late potentials using high resolution or signal - averaged electrocardiography" published in European Heart Journal (1991) 12, pages 473-480 mentions that: "Although there is no definitive evidence yet to support the concept that frequencies above 500 Hz contribute to the differentiation of patient groups, sampling below 1000 Hz may preclude recovery of potential signals of interest". The prior art thus generally considers that in monitoring cardiac activity signals there is no reason to take into account frequency components which are above 500 Hz, since they are considered as not comprising relevant information. The prior art thus clearly leads the skilled person away from analyzing frequency components above 500 Hz.

The present invention now deals with recording and monitoring cardiac activity signals and taking into account frequency components above 500 Hz. For this purpose a device according to the present invention is characterized in that said device comprises signal processing means provided for extracting and processing from said electrical signal those frequency components which are above 500 Hz without removing them. Experience has proven that above 500 Hz relevant information about the cardiac activity could be found. The higher frequency components of the cardiac activity signals enable to provide information at a cellular level and strand of cells or fiber bundles relating to the operation of the heart. The heart tissue can be considered as composed of a plurality of heart cells. All those heart cells contribute to the electrical signal indicative of the cardiac activity. The frequency components below 500 Hz, which are normally considered, represent a smoothed vectorial sum of the signal vectors of each heart cell. On the contrary, the frequency components above 500 Hz represent the contribution of the individual vectors. By considering the higher frequency components, i.e. those above 500 Hz, information of individual heart cells or small bundles is obtained which enables a more accurate prediction of cardiac malfunctioning and consequently an improved therapeutic action in case of disease. In the cited US-A- , 947, 857 high frequency signal are considered as noise which is eliminated. The averaging of the sampled signal is realized to enhance the resolution of the electrocardiogram signals, in particular in the frequency range below 500 Hz. Although figure 6A of US-A-4, 947, 857 illustrates the presence of some signal above 500 Hz, the latter signal is considered as noise and autocorrelated it away. dB compression and autoregression was applied on the recorded signal (figure 7) so that the high frequency signals were filtered out.

A first preferred embodiment of a device according to the invention is characterized in that said processing means are further provided for extracting and processing from said electrical cardiac activity signal said frequency components above 500 Hz and which are present in at least one signal part of the group comprising the P, the QRS, the T and/or the U wave as well as in the intervals between those groups. By monitoring the frequency components above 500 Hz of those waves, information about electrical activity of cardiac cells at a higher resolution compared to standard E.C.G. is obtained.

A second preferred embodiment of a device according to the invention is characterized in that said processing means are provided for extracting from said cardiac activity signal components relating to one of the characteristics duration, amplitude power spectrum, energy content of the cardiac signal. Those signal components provide useful information for diagnostic and therapeutic purpose.

Preferably said processing means are provided for selecting within said cardiac activity signal at least one signal portion relating to a predetermined part of a cardiac cycle. A particular signal portion can thus be selected in order to analyze the electrical signal originating from one particular part of the heart or the cardiac cycle. A third preferred embodiment of a device according to the invention is characterized in that said signal processing means are provided with time window application means for applying on said cardiac activity signal at least one time window within a same cardiac cycle, said time window application means being further provided with comparison means for comparing with a predetermined content the signal content within the time window and for generating a correspondence respectively a non-correspondence signal upon detection of correspondence respectively non-correspondence between said signal content and said predetermined content . By comparing the presence or absence of cardiac activity signals in different time windows it is possible to analyze normal and abnormal cardiac activity. By applying a time window on a particular part of the cardiac activity signal the presence or absence of a signal within that time window is indicative of how the concerned heart operates .

A fourth preferred embodiment of a device according to the invention is characterized in that said time window application means comprises a memory for storing the signal content of at least two sets of said two time windows fetched in different cardiac cycles, said time window application means being provided -for processing for each of said two time windows the signal content of each of said set of time windows. Cardiac disease or cardiac malfunctioning can better be observed by comparing different cardiac cycles with each other.

Preferably said window application means are provided for applying a time shift on each of said two windows in order to time shift each of said windows between subsequent cardiac cycles, said comparison means being further provided for selecting among windows of subsequent cardiac cycles at least the one showing moεc cardiac activity. This enables to scan cardiac activity signals in order to look for the presence or establish the absence of_ a particular signal pattern. A fifth preferred embodiment of a device according to the invention is characterized in that said processing means comprise trend analyzing means provided for comparing signals from subsequent cardiac cycles with each other in order to determine whether said subsequent cycles show a trend. Such trends will help in diagnosing certain diseases such as for example ischemia, or in observing changes in action potential duration or instability of the latter. A sixth preferred embodiment of a device according to the invention is characterized in that said processing means comprise prediction means provided for analyzing said processed cardiac activity signals in order to detect at least one predetermined signal pattern indicative of a cardiac disease. This prediction means are valuable for diagnostic and therapeutic purposes.

Preferably said prediction means are provided for generating a trigger signal upon detection of said pattern. Such a trigger signal can then be used to activate either a pacemaker, a drug dispenser, etc.

A seventh preferred embodiment of a device according to the invention is characterized in that said electrical signal comprises an envelope formed by a low frequency component thereof and representing P, QRS, T and U part and the intervals between those parts, said detecting means being provided for demodulating a further component from said signal which further component is modulated on said envelope. Demodulating the envelope is an alternative manner to obtain the high frequency components of the cardiac activity signals.

Preferably characterised in that said detecting means are provided for applying a time and frequency domain analysis on said further component. The latter analysis is a manner for demodulating the envelope which can be easily implemented in an electronic manner.

The invention also relates to a method for recording and monitoring cardiac activity signals, said method comprises :

- detecting an electrical signal generated by a heart during its activity;

- processing frequency components from said electrical signal;

Such a method is characterized in that said processing is applied on those frequency components which are above 500 Hz without removing them.

Preferably said detecting is realised on a beat- to-beat basis. Beat-to-beat basis enables an accurate tracing of the cardiac activity.

The invention will now be described in more detail with reference to the drawings wherein:

Figure 1 shows an example of an Electrocardiogram (ECG) recorded in a well known manner.

Figure 2 illustrates an intra-cardiac recording.

Figure 3 illustrates schematically the higher frequency components on a typical P, QRS, T and U signal.

Figure 4 shows an example of a device for monitoring cardiac activity signals according to the present invention.

Figure 5 shows the different steps in a signal processed by the device shown in figure 4.

Figure 6 illustrates the application of time windows on the cardiac signal.

Figure 7 shows an amplitude density plot as a function of time and frequency of a Fast Fourier transformed cardiac activity signal obtained from a healthy heart. Figure 8 shows an amplitude density plot as a function of time and frequency of a Fast Fourrier transformed cardiac activity signal obtained from a diseased heart.

Figure 9 shows the recording of a cardiac activity signal such as recorded from the surface in a human being and which signal is processed according to the present invention. In order to obtain information about the functioning of the heart of a patient, the physician will generally record an electrocardiogram (ECG) of which an example is shown in figure 1. For recording the ECG, the bandwidth is usually limited to the low frequency range of 3 to 20 Hz. This signifies that the recorded signal is strongly filtered and thus only yields signals from the larger parts of the heart: i.e. the atria generate the P- wave, depolarisation of the ventricles produces the QRS- complex, while repolarization is seen as a T-wave. The important electrical activity of some of the smaller parts of the heart such as His bundle activation and sinus node activity, is not distinguished in the signal, but the physician can derive much of that cardiac activity indirectly. Thus the sinus node activity precedes the P- wave, while conduction through the AV-node is reflected in the time between the onset of the P-wave and the QRS complex.

A method to obtain more information about those smaller parts of the heart uses intra-cardiac recording. For this purpose recording electrodes are placed in the heart . For example by introducing an electrode over the His bundle, the activity of the impulse going through this small bundle can be directly measured, enabling the physician to subdivide the activation delay between the atria and the ventricles in two parts, i.e. the atrium to

His (A to H of the cycle) and His to ventricle ( H to V part of the cycle) conduction times as shown in figure 2.

The cardiac activity signals recorded in the ECG or with intra-cardiac electrodes are in fact a summation of the activity of many millions of individual cardiac cells. The heart is constituted of a large number of individual cells which are grouped into fibers, each cell upon activation develops a vector whose signal contributes to build up the summation. Since the ECG is a summation of the activity of all cardiac cells, the ECG can not provide an information about individual cells. Thereupon the electrical signals from each individual cell are so small and so fast (depolarisation lasts on the order of 1 msec) , that only a summation can be observed on the surface ECG. At a bandwidth of 3-20 Hz the QRS-complex is observed. Further increase of the bandwidth leads to the observation of some notches originating from the activation of small cardiac fibers..

In the known cardiac activity signal monitoring, the signals originating from the heart and having a frequency above 500 Hz are generally considered as irrelevant since they are considered as bringing no contribution to the differentiation of patient groups. The latter assumption is mentioned in the article "Standards for analysis of ventricular late potentials using high resolution or signal-averaged electrocardiography" published in European Heart Journal (1991) 12, pages 473- 480.

Contrarily to the latter assumption, the method according to the present invention takes into account the cardiac activity signals having a frequency above 500 Hz. According to the present invention a device for recording and monitoring cardiac activity signals has been developed, which device comprises signal processing means provided for extracting and processing from at least a part of the electrical signal generated by a heart during its activity, those frequency components which are above 500 Hz. The extracted frequency components are not removed form the signal but subjected to a signal processing and analyzing which will be described hereunder. Starting from the principle that ECG and other conventional recording only considers a summation signal whose frequenc - components are situated within a frequency range below 500 Hz, the processing of the signal having a frequency above 500 Hz will enable the detection of a signal originating from small cardiac fibers and in particular to make an attempt to observe signals originating from individual cells or at least from a limited number of individual cells .

The observation of cardiac activity signals having a frequency above 500 Hz could be compared to microscopic observation. Upon looking to a small frequency range of for example 3 to 20 Hz only a large general structure can be observed. Just as by increasing the magnification factor of a microscope, smaller details become more visible, likewise broadening the frequency bandwidth of the device resolves smaller and faster components composing the cardiac signal.

The more detailed observation of cardiac activity which becomes possible by application of the method and the device according to the present invention is illustrated in the figures 7 and 8. Figure 7 respectively 8 shows an amplitude density plot as a function of time and frequency of a Fast Fourrier transformed cardiac activity signal obtained from a healthy heart respectively a diseased heart . These amplitude density plots are obtained by processing the cardiac signal according to the present invention. As can be seen from figure 7 the healthy heart shows clearly some activity above 500 Hz, and this activity shows a well organised pattern. On the contrary, the diseased heart

(figure 8) shows a lot of activity above 500 Hz. However, the maximal amplitude of the latter activity is lower than the one of the healthy heart and the cardiac activity is less focused within a small frequency range. Analyzing the cardiac activity signals above 500 Hz can thus provide a lot of information about how the heart functions and in particular it can help detecting cardiac disease or provide some information which will help to prevent that a disease would have fatal consequences.

Figure 3 illustrates a typical P, QRS, T and U complex, upon which higher frequency components originating from small tissue fibers have been superposed. The higher frequency components could be considered as being modulated on the lower frequency signal . Removing of the lower frequency components would then be obtained by demodulating of the signal. Such a demodulation operation then provides frequency components which are directly related to electrical heart activity at a cellular level. As illustrated in figures 7 and 8 high frequency components are indeed present in the electrical signal produced by the heart during its activity. Those high frequency components should for a healthy heart normally be present in the P, QRS, T and U complex as these complexes are representative for well defined cardiac activity.

Figure 9 shows an amplitude versus time diagram obtained by recording the absolute value of cardiac activation signals. The signals were recorded by using two precordial surface leads non-invasively applied on a human volunteer. Electronically all recording was made with a high bandwidth (100 KHz) amplifier. The amplified signal was sampled at 300 KHz. Cardiac activity is observed in the latter diagram notwithstanding the use of precordial surface leads, which signifies that by use of the present invention detailed information about cardiac activity can be obtained without invasive recording. The diagram of figure 9 illustrates that in synchronism with each activation of the heart there is a large transient increase in high frequency cardiac activation signals.

The power of the higher frequency components of the electrical signals generated by a heart during its activity is much lower than the power of the lower frequency components . Further those higher frequency components may be hidden in the background noise of the signal if proper noise reduction is not applied. In order to distinguish those higher frequency components an electronic processing of the electrical signal will be necessary. Figure 4 shows a schematic view of an example of a device for monitoring cardiac activity signals according to the present invention. The latter device comprises a sampling unit 4 to which the electrical signal generated by the heart and picked up by electrodes 1, 2 and 3 is supplied. The electrodes can be as well intra-cardiac electrodes as surface electrodes. The number of electrodes is not relevant for the present invention and will be determined in function of the diagnostic to be made and/or the signals to be measured. The sampling unit is provided for sampling at a fast sampling frequency ( several KHz) , the signals supplied by the electrodes. The sampling unit 4 is connected with a buffer 5 which is provided for temporarily storing the recorded samples in order to enable a later processing.

An output of the buffer 5 is connected with an input of a bandpass filter unit 6 provided for extracting from the electrical signal those frequency components which are above 500 Hz. The filtered signal samples are then supplied to a differentiation unit 7 and thereafter to an absolute value transformation unit 8, whose output is connected with an activity detector 9. The filter unit 6 is preferably a high pass filter provided for extracting the high frequency components of the QRS complex and a low-pass filter for preventing aliasing. The filter unit is operative on frequency components which are present in the P, the QRS, the T and the U wave.

Figure 4 only shows a preferred embodiment of a device according to the present invention. Alternative embodiments are of course possible. So for example the filter unit 6 could be followed by an Analog / Digital convertor and a Fourrier transformation unit instead of applying a differentiation (as was actually done for obtaining the result shown in figures 7 and 8) . The filter could also be a narrow band pass filter followed by an absolute value determination unit . The Fourrier transformation unit will then apply a time and frequency domain analysis on the signal .

The operation of the device shown in figure 4 will now be described. The electrical signal generated by the heart during its activity is picked up by electrodes 1, 2 and 3 and transferred to the sampling unit which samples the electrical signal and the samples thus obtained are temporarily stored in the buffer 5. The samples are then filtered in order to extract those frequency components which are above 500 Hz. Figure 5 shows in part A the output of the total signal (high and low frequency components) such as output by buffer 5. That total signal comprises the content of the upstroke of the action potential from the electrical signal . The upstroke of the action potential can already clearly be observed in that part A of the cardiac activity signal. Similar CAS is observed in the down going part due to the T wave . The high frequency components, indicative of smaller cardiac fibers can however still not be observed from this processed signal since they are still below the resolution of the trace .

In figure 5 the time (t) respectively the amplitude (a) is set out in the horizontal respectively vertical direction. Signal c indicates the signal originating from the pressure developed by the heart . Signal p respectively m indicates the signal originating from the Purkinje fibers respectively the myocardial fibers. The repolarization of the T wave is observed on the 200 to 300 ms time period following the activation.

In order to extract cardiac activity signals from the upstroke of the cardiac action potential a time differentiation is applied on the filtered signal by means of the differentiation unit 7. The differentiation is applied separately on the signal originating from the Purkinje and from the myocardial fibers. Figure 5C shows those differentiated signals.

Instead of applying a differentiation on the signal a Fast Fourrier transformation could be applied on the power spectrum of the signal .

The absolute value transformation unit 8 now applies the following operation on the signal.

∑ I d _jl + I dm_j dt dt wherein dp_i/dt respectively dm-_j/dt indicates samples at a same moment in time in the Purkinje respectively myocardial signal .

After determination and addition of the absolute values, the latter are fed to the activity detector 9. This detector is provided for comparing a supplied signal to a predetermined threshold which is for example a statistically determined threshold. The onset of cardiac activation t₀ is determined when the signal input in the activity detector 9 first exceeds the predetermined threshold value. The end of the cardiac activation t- is determined when the supplied signal remains below that threshold value. The signal output by the activity detector 9 is illustrated in figure 5B. That signal also indicates how much the input signal exceeds the threshold value in such a manner that the amplitude of the output signal in indicative of the amplitude of the cardiac signal. From that output signal the duration t- - t₀ of the cardiac activity during one cycle can now be obtained.

The monitoring of the cardiac activity by means of the device according to the present invention can be realised in different manners of which some will be described hereunder by way of example. The whole cardiac signal can be monitored or only one or more signal portions relating to a predetermined part of a cardiac cycle. When only a portion is monitored the relevant portion as for example selected on a time basis, triggered by the stimulus, and only a limited number of samples representative of the considered time period are taken into account . The cardiac signals are sampled either beat by beat or over a series of subsequent heart beats. Instead of picking up the cardiac signal on a single location, the signal from different locations of the heart could be monitored. The device according to a preferred embodiment is provided with time window application means for applying on said filtered cardiac signal at least one time window within a same cardiac cycle. The time window application means (not shown in figure 4) are located between the filter unit and the differentiation unit. The application of time windows W_lf W₂, ... is illustrated in figure 6. The time window application means for example comprise a P A or a microprocessor with local memory. The time window application means are provided either to apply a single time window within a same cycle or to apply two or more time windows within a same cycle . In the latter case the time windows are preferably non-overlapping The time windows are preferably applied on those segments of the cardiac signal where the presence or absence of a signal is indicative for the functioning of the heart . So the time windows could be applied on the segment where the

P, the QRS, the T or U complex is expected or in between.

The time window application means are further provided with comparison means for comparing the content of one or more windows with a respective predetermined content in order to verify whether the content of the concerned time window corresponds with that predetermined content. Upon non-correspondence a trigger signal is than generated. The correspondence check should not necessarily be a one-to-one correspondence but merely a check whether the signal has a predetermined amplitude and direction.

Instead of comparing one time window content with a predetermined content, the content of one or more time windows could also be analyzed in order to verify the presence or absence of a predetermined signal. Indeed the absence of a particular signal component in a particular time window, or its presence in a time window where it is not expected, indicates a cardiac disease. A search operation could also be applied on the different time windows in order to search where a particular signal component is located. The comparison could also take place by comparing two time windows with each other and verify whether they show a similar content or not.

Beside a comparison operation, the time window application means can also be provided with a memory for storing the signal content of at least two subsequent sets of two time windows fetched in subsequent cardiac cycles. Said contents are then compared with each other in order to verify whether the cardiac signal remains stable or not.

Instead of applying the time window(s) each time on a same time period of the signal, the time window(s) could also be time shifted over the signal between subsequent cardiac cycles or within a same cardiac cycle. In the latter case the comparison means are further provided for selecting among windows of subsequent cardiac cycles at least one showing a predetermined cardiac activity. The latter is for example realised by each time comparing on the signal content of the time shifted window and selecting the one comprising the predetermined activity.

The comparison of the signal content of a same window for subsequent cardiac cycles is a further preferred option of the comparison means. By comparing subsequent samples within a same window, trends in the cardiac signals could be observed. Those trends, or even a particular shape in a single signal provide information which enables to predict a cardiac disease. Upon detection of such a trend indicating a cardiac disease a trigger signal is then generated. Such a trigger signal is issued by a generator cooperating with said comparison means. The trigger signal could then activate a pace maker, a defibrillator, a drug pump or the like.

It should be noced that Cardiac Activity Signals are not only associated with depolarization. Indeed upon repolarization there are similarly electrical vectors between cells that do not repolarize in synchrony. These vectors have a lower amplitude and a lower frequency content than those of depolarization. Those repolarization signals are also monitored with the device according to the present invention. Such repolarization CAS is a powerful indicator of the asynchrony of repolarization or its beat-to-beat instability, which in turn predicts the development of repolarization disturbances (e.g., EADs (Early After Depolarization) DADs (Delayed After Depolarisation) etc) and thereby forewarn impending serious arrhythmias (e.g., torsade de pointes, ventricular tachycardia or fibrillation) .

The cardiac activity signals monitored according to the present invention could be used for several purposes. CAS signals could be used for atrial or ventricular capture during pacing (e.g., automatic verification of the paced evoked response) . Suppose that a pacemaker produces during 0,5 ms a stimulus with a 5 Volt amplitude. If the heart reacts on the pacing stimulus, then cardiac activity signals are produced which are then monitored with the device according to the present invention. Capture of the pacemaker stimulus by the heart can thus be monitored. The presence or absence of cardiac activity signals enables to automatically verify whether or not a pacing stimulus is followed by an atrial or ventricular depolarisation.

Threshold (voltage and time) determination or tracking for pacemaker stimuli could also be realized according to the present invention. Suppose that the pacemaker is adjusted on 5 V, 0, 5 ms stimuli, while a 0,7 V, 0,5 ms stimulus could be enough. By ratiometric comparison of the interval with the expected CAS having a "blank" interval the threshold of the amplitude of the pulse can be determined. Detecting, according to the present invention whether or not a pacing stimulus results in atrial or ventricular capture enables to determine the minimum threshold for capture.An energy saving and consequently a substantial longer lifetime of the pacemaker is thus obtained. Each time adjusting (decrease or increase) the amplitude and/or the duration of the stimulus and checking the effect on the CAS an adjustment of the threshold is realised.

The automatic determination of threshold by analyzing the cardiac activity signals monitored according to the present invention, enables the delivery of trains of subthreshold stimuli to prevent the occurrence of atrial and ventricular tachycardias, or to terminate atrial and ventricular tachycardias.

Cardiac activity signals enable to differentiate between stimulus artefacts and intrinsic cardiac signals. This may prevent auto-inhibition of the pacemaker by sensing the stimulus artefact.

Cardiac activity signals enable to discriminate between intrinsic cardiac activity and electromagnetic interference, which may prevent pacemaker inhibition by noncardiac signals or electromagnetic interference. Oversensing of pacemaker stimuli by an implantable cardioverter-defibrillator (ICD) may lead to double or triple counting of the heart rate by sensing the atrial or ventricular pacemaker stimulus. This double or triple counting may exceed the cutoff rate of the ICD and may initiate an inappropriate discharge of the ICD, even though the rhythm is neither ventricular tachycardia or ventricular fibrillation. This may be prevented by CAS as it discriminates between cardiac depolarisations and pacemaker stimuli . Differentiation between AV-nodal reentry tachycardia and atrioventricular tachycardia with accessory pathway including its termination is another possible use of the CAS. An AT nodale retro tachycardia due to the presence of a Wolf-Parkinson-White contraction in the heart would clearly be observed in the CAS. The CAS would also enable to observe a differentiation between stimulus artefact and intrinsic cardiac signals (e.g., enabling monitoring atrial signals) , and thus prevent auto inhibition of the pacemaker.

After the delivery of a shock by an ICD the detection of cardiac activity is extremely difficult. Cardiac activity signals monitored according to the present invention allows the detection of the absence or presence of intrinsic cardiac signals and may trigger antibradycardia pacing when needed.

The interval between atrial or ventricular stimulus to peak repolarisation cardiac activity signals monitored according to the present invention can be used in a pacemaker as a measure of the metabolic needs of the patient during physical exercises or emotional stress to adapt the pacing frequency of the pacemaker in a physiologic way.

The time interval between atrial cardiac activity signals monitored according to the present invention and the ventricular cardiac activity signals monitored according to the present invention is another method to evaluate the adrenergic tone of a patient and can be used as a measure to adjust the pacemaker frequency to the physiologic needs of the patient. Rate responsive pacing thus becomes possible.

Analysis of the atrial cardiac activity signals monitored according to the present invention allows the differentiation between anterogradely or retrogradely conducted P waves. Recognition of retrogradely conducted P waves permits the detection of a pacemaker mediated tachycardia in dual chamber pacemakers (such as DDD(R) , VDD(R) , VAT pacemakers) . Upon detection of a Pacemaker Mediated Tachycardias (PMT) an adequate termination stimulation protocol can be elicited.

Atrial cardiac activity signals monitored according to the present invention can be used to detect retrograde atrial activation and can measure the retrograde ventriculo-atrial (V-A) interval. Upon recognition of a prolonged V-A interval the pacemaker can then be provided for adjusting its post ventricular refractory period (PVRARP) to prevent the occurrence of PMTs.

Atrial cardiac activity signals monitored according to the present invention can detect atrial tachyarrhythmias during dual chamber pacing. To avoid tracking of fast atrial rates cardiac activity signals monitored according to the present invention can control the algorithm which switches the pacemaker to a nontracking mode (such as VDIR, WIR, DDIR mode) .

The relative time intervals between the atrial and ventricular cardiac activity signals monitored according to the present invention as well as the discrimination between retrogradely and anterogradely conduction allows to differentiate between an atrial tachycardia (AT) , an AV nodal reentry tachycardia (AVNRT) and an atrioventricular tachycardia (AVRT) with an accessory pathway (AP) with either fast or slow retrograde conduction properties. Upon recognition of the tachyarrhythmia an appropriate pacing termination protocol can be initiated.

Analysis of atrial and ventricular cardiac activity signals monitored according to the present invention may differentiate between supraventricular and ventricular tachycardias and control the algorithm for termination of those arrhythmias as well in pacemakers (PMs) as in implantable cardioverterdefibrillators (ICDs) . It further may prevent the delivery of inappropriate shocks in ICDs. Recognition of bundle branch block and differentiation between supraventricular tachycardia with aberrant conduction and ventricular tachycardia is thus possible .

Atrial cardiac activity signals monitored according to the present invention enables the detection of atrial flutter (A.FLUT.) and atrial fibrillation (A. FIB.) . Upon recognition of the tachycardia controlling the algorithm in order to terminate these arrhythmias either by overdrive pacing (for A.FLUT.) or by delivering subthreshold stimuli for atrial flutter or by delivering shock therapy in implantable atrial cardioverter¬ defibrillators is possible. Ventricular cardiac activity signals monitored according to the present invention show a specific pattern at the occurrence of ventricular tachycardia and ventricular fibrillation and can be used to initiate a tiered therapy such as decremental pacing (for VT) or delivery of low-energy and high-energy shocks by an ICD.

Cardiac activity signals monitored according to the present invention become unstable prior to the occurrence of ventricular fibrillation (VF) . This allows to take measures to prevent ventricular fibrillation. Cardiac activity signals monitored according to the present invention will allow the recognition of different tachyarrhythmias and can be used in an ECG machine for automatic analysis of electrocardiograms.

Cardiac activity signals monitored according to the present invention signals can measure QT dispersion and detect as well early afterdepolarisations (EADs) as delayed afterdepolarisations (DADs) and episodes of torsade de pointes (TdP) . Upon detection antiarrhythmic measures can be taken to prevent or terminate realted rhythm disorders.

Cardiac activity signals monitored according to the present invention can detect late potentials and this can easily be incorporated in an ECG machine.

Cardiac activity signals monitored according to the present invention allows an accurate measurement of intracardiac conduction intervals and may control the delivery of an appropriate drug or electrical therapy.

Cardiac activity signals monitored according to the present invention show alterations at the onset of ischemia and can differentiate between viable and non- viable myocardium. This can be used to control drug delivery by an automatic drug dispenser or to start an electrical therapy.

Cardiac activity signals monitored according to the present invention can differentiate between asystole and e.g. fine ventricular fibrillation (VF) . This detection system can be incorporated in cardiac monitoring systems and in automatic implantable and external defibrillators.

The presence of electrical activity without mechanical contraction (electro-mechanical dissociation or pulseless activity) can be recognized by cardiac activity signals monitored according to the present invention. This can be implemented in cardiac monitoring systems on intensive care units and in automatic implantable and external cardioverter-defibrillators (ICDs) .

Cardiac activity signals monitored according to the present invention can detect atrial and ventricular premature beats and can control the algorithm either to prevent or to treat these arrhythmias .

Cardiac activity signals monitored according to the present invention can be used during electrophysiologic (EP) studies as it can determine maximum follow frequencies and refractory periods, verify capture, measure intracardiac conduction intervals, recognize antegrade and retrograde conduction and different arrhythmias and can localize the origin of arrhythmias and pre-excitation in patients with the Wolff- Parkinson-White (WPW) syndrome in order to guide radiofrequency (RF) ablation procedures.

Arrhythmia recognition by cardiac activity signals monitored according to the present invention through paddles of an external defibrillator may be of value for e.g. paramedics using the device in emergency medicine and rescue teams .

Cardiac activity signals monitored according to the present invention can discriminate between artefacts and cardiac activity and thereby reduce false alarms during cardiac ECG monitoring and telemetry.

Cardiac activity signals monitored according to the present invention allows accurate transmission of electrocardiographic data via programmer heads, telephone or other telemetry means (e.g. GSM) for regular follow-up or for surveillance for patients at risk for life- threatening rhythm disorders.

Recording of cardiac activity signals monitored according to the present invention via external leads may supplement standard ECG recordings and may be helpful for diagnostic and therapeutic purposes.

Claims

1. A device for recording and monitoring cardiac activity signals, said device comprises detecting means provided for detecting an electrical cardiac activity signal generated by a heart during its activity, characterized in that said device comprises signal processing means provided for extracting and processing from said electrical cardiac activity signal those frequency components which are above 500 Hz without removing them.

2. A device as claimed in claim 1 characterized in that said processing means are further provided for extracting and processing from said electrical cardiac activity signal said frequency components above 500 Hz and which are present in at least one signal part of the group comprising the P, the QRS, the T and/or the U wave as well as in the intervals between those groups .

3. A device as claimed in claim 1 or 2 characterized in that said processing means are provided for extracting from said cardiac activity signal components relating to one of the characteristics duration, amplitude power spectrum, energy content of the cardiac activity signal.

4. A device as claimed in claim 3, characterized in that said signal processing means are further provided for applying a Fourier transformation on said spectrum for extracting frequency signals within the range of 0.25 Khz to 10 Khz.

5. A device as claimed in claim 3 or 4, characterized in that said processing means are provided for selecting within said electrical cardiac activity signal at least one signal portion relating to a predetermined part of a cardiac cycle.

6. A device as claimed in claim 5, characterized in that said processing means provided for performing said selection on a time basis.

7. A device as claimed in any one of the claims 3 to 6, characterized in that said signal processing means are provided with time window application means for applying on said cardiac activity signal at least one time window within a same cardiac cycle, said time window application means being further provided with comparison means for comparing with a predetermined content the signal content within the time window and for generating a correspondence respectively a non-correspondence signal upon detection of correspondence respectively non- correspondence between said signal content and said predetermined content.

8. A device as claimed in claim 7, characterized in that said time window application means comprises a memory for storing the signal content of at least two sets of said two time windows fetched in different cardiac cycles, said time window application means being provided for processing for each of said two time windows the signal content of each of said set of time windows.

9. A device as claimed in claim 7 or 8, characterized in that said window application means are provided for applying a time shift on each of said two windows in order to time shift each of said windows between subsequent cardiac cycles, said comparison means being further provided for selecting among windows of subsequent cardiac cycles at least one showing cardiac activity.

10. A device as claimed in any one of the claims 3 to 6, characterised in that said processing means are provided with time window application means for applying on said cardiac activity signal a time window, said time window application means being provided for applying a time shift on said window in order to shift said window within said cardiac signal.

11. A device as claimed in any claim of the claims 1 to 10, characterized in that said processing means comprises a low pass filter provided for preventing aliasing.

12. A device as claimed in any one of the claims 1 to 11, characterized in that said processing means comprise trend analyzing means provided for comparing signals from subsequent cardiac cycles with each other in order to determine whether said subsequent cycles show a trend.

13. A device as claimed in any one of the claims 1 to 12, characterized in that said processing means comprise prediction means provided for analyzing said processed cardiac activity signals in order to detect at least one predetermined signal pattern indicative of a cardiac disease.

14. A device as claimed in claim 13, characterized in that said prediction means are provided for generating a trigger signal upon detection of said pattern.

15. A device for recording and monitoring cardiac activity signals, said device comprises detecting means provided for detecting an electrical cardiac activity signal generated by a heart during its activity, characterized in that said electrical signal comprises an envelope formed by a frequency component less than 500 Hz thereof and representing P, QRS, T and U part and the intervals between those parts, said detecting means being provided for demodulating a further component above 500 Hz from said signal which further component is modulated on said envelope.

16. A device as claimed in claim 15, characterised in that said detecting means are provided for applying a time and frequency domain analysis on said further component .

17. A method for recording and monitoring cardiac activity signals, said method comprises : detecting an electrical cardiac activity signal generated by a heart during its activity; - processing frequency components from said electrical cardiac activity signal; characterized in that said processing is applied on those frequency components which are above 500 Hz without removing them.

18. A method as claimed in claim 17, characterized in that said detecting is realised on a beat-to-beat basis.

19. A method as claimed in claim 17 or 18, characterized in that said electrical cardiac activity signal is sampled at a sampling frequency situated in a range of 500 Hz to 100 KHz.

20. A method as claimed in any one of claims 17 to 19, characterized is that said detecting is realised at different times of a same cardiac cycle.

21. A method as claimed in- any one of the claims 17 to 21, characterized in that said electrical cardiac activity signal is fetched on a plurality of locations in the heart .

22. A method as claimed in any one of the claims

17 to 21, characterized in that said method further comprises an analysis of the frequency components above 500 Hz in order to detect a heart disease.

23. A method as claimed in claim 22, characterized in that said detection of heart disease comprise the detection of one of the disease ischemia, onset of fibrillation, onset of tachycardia.

24. A method as claimed in any one of the claims 17 to 23 characterized in that said processed cardiac signal is compared to threshold values and upon exceeding said threshold value an energy threshold signal for a pacemaker is generated.